Description (provided by application): The last decade has brought revolutionary new techniques allowing visualization of the working brain in humans at the systems level. However, a large gap remains between the spatiotemporal resolution of tomographic techniques (fMRI, PET), and the circuit level where animal studies permit mechanistic neural models. It is the overall goal of this proposal to develop an integrated suite of technologies to bridge this critical gap. Two interrelated themes are found throughout this proposal: (1) to improve the spatial and temporal resolution of non-invasive technologies, which will enable direct imaging of discrete (e g column and laminar level) neural units which bridge the systems and cellular levels and (2) to clarify the mechanisms which relate the biophysics of neuronal activity "observables" in our imaging measurements. The two key technologies to be investigated are: (1) extremely high resolution MRI and fMRI, using very high strength gradients, phased-array coils, and other advances at 3T and 7T non-human primates, and 9.4T rats and (2) tomographic optical imaging, increasing the resolution and physiological range using three different optical technologies: direct reflectance imaging, optical scanning microscopy, and diffuse optical tomography. These technologies will be validated against invasive "gold standard" techniques in studies of rat whisker barrel cortex and macaque visual cortex, and further applied to animal models spreading depression in migraine and stroke. Each of these experiments is designed to allow us to serially step from more to less invasive, and move from systems where much is already known through to studies in humans that have not heretofore been explored within the spatiotemporal domains our newly developed tools will afford.